Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Go ahead, write the “stupid” code

spikepuppet.io

Post reply on HN
Print view View original

1–10 of 157 posts

Re: Go ahead, write the “stupid” code

#3

Post by anon-3988 »

I feel like people should be writing stupid code, and in the case where its a compiled language, we should ask compiler or the language for better optimization. The other day, I was writing a check of a struct that have certain structures (protobuf probably have something like this)

struct S { int a; int b; int c; int d; int e; /* about 15 more members */ }

so I wrote

const auto match_a = s.a == 10; const auto match_b = s.c == 20; const auto match_c = s.e == 30; /* about 15 more of these */ if (match_a && match_b && match_c) { return -1; }

Turns out compilers (I think because of the language) totally shit the bed at this. It generates a chain of 20 if-else instead of a mask using SIMD or whatever. I KNOW this is possible, so I asked an LLM, it was able to produce said code that uses SIMD.

Re: Go ahead, write the “stupid” code

#4

Post by bitwize »

I'm working on in-kernel ext3/4fs journalling support for NetBSD. The code is hot garbage but I love it because of the learning journey it's taken me on: about working in a kernel, about filesystems, etc. I'm gonna clean it up massively once I've figured out how to make the support complete, and even then I expect to be raked over the coals by the NetBSD devs for my code quality. On top of that there's the fact that real ones use ZFS or btrfs these days, and ext4 is a toy; like FAT, by comparison, so this may not even be that useful. But it's fun and lets me say hey Ma, I'm a kernel hacker now!
Post reply on HN